THE BUTTERFLIES OF CEYLON 101 
very slightly concave, if at all. It has a very broad black 
border, with only a few minute, or no, blue spots on it. 
and, spiculifera.—Smalier and of a paler blue. The 
termen of the fore wing more falcate, and the hind wing 
proportionately smaller. The black border of the fore wing 
is much narrower, especially near the tornus, and there is a 
series of small elongated blue spots on this border, that in 
interspace 3 being shifted inward, and the two in interspace I 
almost or quite joined to the ground colour. As Moore points 
out, ceylonica appears to be nearest to pingasa and spiculifera 
tohippia. His variety fraterna appears to be an intermediate 
form. 
Ceylonica is common in the drier low-country, but is found 
at all elevations during the flights, but I have only taken 
spicultfera in the hills. Both are variable, and it is easy to 
grade them. 
Personally I believe spiculifera to be the cold season (or high 
elevation) form, while /raterna and ceylonica are the dry and 
wet season forms respectively of those bred in the low-country. 
Indian writers agree in regarding pingasa and hippia as distinct 
species, and it is curious that varieties of our Ceylon race 
should almost grade into each of them. An analogous 
case is that of Tevias leta and venata, the former of which is 
allied to the Ceylon T. rama and the latter to T. cingala, 
and yet there can be little doubt that rama and cingala are 
only varieties of our Ceylon race. 
The females vary greatly in the width of the blue markings 
on the upper side and the amount of black shading below. 
In June, 1916 and 1917, I got specimens at Wellawaya, which 
were very exceptionally dark both above and below; in fact, 
all flying there then were of this unusual colour. The males 
with them seemed to be normal specimens of ceylonica above, 
but had more black shading below. I have received a similar 
pair from Kandy. These dark varieties of the female are 
good mimics of D. septentrionis. Normal varieties mimic D. 
limntace and aglea, but especially resemble D. exprompta, 
which, however, is not found in the usual haunts of Pareronia. 
